Wednesday, October 01, 2003

I feel like I'm 2 keywords shy of a good google search on most of these issues.  After spending the evening reading through Bill Harlan to earn enough karma to ask: Microsoft Natural keyboards are loud. Headphone penetrating focus thwarting loud. Any suggestions? Is there a definitive regular expression for UR[L|I] address matching? Do you have any suggestions or resources for learning large chunks of (plain) code? Does your approach differ if it's a. pattern peppered OO, and b. of it is on top of partially and to-be-learned commercial and proprietary libraries? Is using a dial-up Windows XP (home) box as internet gateway (or DHCP?) for getting my future Linux setup internet connected (similar to my current Win98) unreasonable? I feel like I'm 2 keywords shy of a good google search on most of these issues. I don't know if this is that Bill Harlan, but I found it amusing.  Aahz's sig. "'It is easier to optimize correct code than to correct optimized code.' --Bill Harlan" brought me to my this investigation 10:51:37 PM     CP4E get's regularly noted when folks talk about expanding Pythons m???share by commencing more programmers. Similar to the "create, don't overtake, markets" thought. Mats Wichmann threw this discussion an appropriate curveball by asking: "If Python is so easy for humans to program in, shouldn't it also be an easy language for programs to program in ...?" Fredric Lundh addresses the literal issue here. I have never analyzed code generation from UML tools before, but a comparison of model to code from a C++, Java and Python example would add insight here, although the tip of the iceburg for Mats thoughts.  Jeff Sutherland threatens: "Therefore, American programmers must find a way to be ten times as productive or they are history", following it up with the dire need for Model Driven Development and resources.  Hmmmm... Python & Java: a Side-by-Side Comparison opens with "A programmer can be significantly more productive in Python than in Java. How much more productive? The most widely accepted estimate is 5-10 times".  Wizards fall in there somewhere also, but are just one slice of that taxonomy. I like task automation wizards and that let me save the steps for later use, ideally in a plain text editable fashion. Which leads to... Embedded Python for "code that pushes applications" is also an area where python could use more face time. VBScript is a tired poster-boy, and it seems that with the latest OpenOffice/STAR Office, the window of opportunity is being addressed with Python-UNO. Also, of note: Learning Tree's Python training is no longer supported! Many alternatives exist, including this one during October in Colorado by a University of Wisconsin graduate. 2:20:42 AM

© Copyright 2003 Dean Goodmanson. Last update: 10/30/2003; 10:27:16 PM.